From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-vc0-f174.google.com (mail-vc0-f174.google.com [209.85.220.174]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id DE3B421F19B for ; Wed, 19 Dec 2012 04:25:51 -0800 (PST) Received: by mail-vc0-f174.google.com with SMTP id d16so2138195vcd.19 for ; Wed, 19 Dec 2012 04:25:50 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=0E355Qqxwv4wS7R+2x0Yu8qtT+gsEzIAMSn/Fg06aPs=; b=ALUh2fc68H7DF6AX3EuAp7FL+YEsMcxHwDAJDvwomzADOntS9fKBZlBB35vWpCHXZF fK3gdMnS3ADSCCLfnebj1nXe6WsmwmgJIk1i3kP5LfUvHImmXrG9uIxzKcaQvRVryqh7 AM0Rv113cUt2UUrb+JdRMMUeNemA2JKlWI25RL7TdZYE0Tb1n5z4q0ursz3bzoKOYHic Dp/mZoOOBcvM3/qU5R6eRDsSgI39/jO8I62Nfkmn7FCgu6YBWNJj9RATzamKomjS33R6 ino7YmPxJFHLO8jhdKqA3p073sMvEPq4mdLkceVhmGT0bW7fCJLkAl5dBy2/C3mUMzdR iK0Q== Received: by 10.52.89.106 with SMTP id bn10mr7407357vdb.68.1355919950466; Wed, 19 Dec 2012 04:25:50 -0800 (PST) MIME-Version: 1.0 Received: by 10.59.8.130 with HTTP; Wed, 19 Dec 2012 04:25:35 -0800 (PST) In-Reply-To: References: <50D19D6B.3000809@cs.rwth-aachen.de> From: Jesper Louis Andersen Date: Wed, 19 Dec 2012 13:25:35 +0100 Message-ID: To: Dave Taht Content-Type: multipart/alternative; boundary=20cf307f3a56fdcebd04d133b66b Cc: bloat Subject: Re: [Bloat] Fwd: [Ns-developers] BitTorrent for ns-3: Call for Code Review X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 19 Dec 2012 12:25:52 -0000 --20cf307f3a56fdcebd04d133b66b Content-Type: text/plain; charset=UTF-8 On Wed, Dec 19, 2012 at 1:09 PM, Dave Taht wrote: > I would like very much to be able to analyze stuff like fq_codel under > bit torrent like loads. > I have some knowledge in this area which may help. You should be aware that BitTorrent may use uTP which is a TCP alternative that tries to eliminate bufferbloat. The main difference is that uTP targets a "full line" delay and adjusts its window size based not only on packet loss, but also on delay. Line delay is then forwarded back over the ACKs. uTP runs inside UDP. You may want to analyze both things. I have a hunch fq_codel solves most of the trouble they had because it eliminates the otherwise standing queue. So be sure to test TCP-only client models as well as the uTP one. -- J. --20cf307f3a56fdcebd04d133b66b Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


On Wed, Dec 19, 2012 at 1:09 PM, Dave Taht <dave.taht@gmail.com&= gt; wrote:
I would like very much to be= able to analyze stuff like fq_codel under
bit torrent like loads.

I have some knowled= ge in this area which may help. You should be aware that BitTorrent may use= uTP which is a TCP alternative that tries to eliminate bufferbloat. The ma= in difference is that uTP targets a "full line" delay and adjusts= its window size based not only on packet loss, but also on delay. Line del= ay is then forwarded back over the ACKs. uTP runs inside UDP.

You may wan= t to analyze both things. I have a hunch fq_codel solves most of the troubl= e they had because it eliminates the otherwise standing queue. So be sure t= o test TCP-only client models as well as the uTP one.


--
J.
--20cf307f3a56fdcebd04d133b66b--